A comprehensive study of domain switching process in different ferroelastic and ferroelastic/ferroelectric perovskite structured ceramics has been performed. The effects of thermal fluctuations on domain switching dynamics were investigated in the ferroelastic and in the ferroelectric case under static and dynamic electric and mechanical conditions. In the ferroelastic case, domain switching behaviour was investigated for different compositions, using different types of mechanical tests. Compression tests were carried out to characterize the ferroelastic properties, such as coercive stress, hysteresis loop and irreversible strain. Creep experiments were performed to study the domain switching time dependence at different stress levels. Domain switching kinetics during creep was characterized by implementing a rate model, based on thermal activation rate theory, which allowed the activation volume to be estimated. A Rayleigh-type analysis was performed to study the effects of stress amplitude, loading rate, temperature and composition on ferroelastic switching. Rayleigh-type relationships were proposed to fit the results and the rate model developed was applied to quantify the effect of the loading rate on the Rayleigh loops. Alternative methodologies were developed to assess the effects of rate and temperature on the coercive stress, providing original sets of data. A further application of the rate model provided an estimation of the activation parameters (volume and enthalpy). In PZT 5A at the coercive field the activation volume was calculated to be 2.44 nm3, with a reasonable consistency with the value obtained from creep tests (7.49 nm3). In the ferroelectric case, domain switching was studied by generating P-E and butterfly hysteresis loops and by analysing creep-relaxation curves. In creep experiments, the polarization and the strain were measured simultaneously, during the application of a constant electric field. An insight into the evolution of domain structure and on domain switching mechanisms was gained, highlighting analogies and differences with the ferroelastic case. Experiments at different frequencies, allowed the activation volume to be estimated at the coercive field (77 nm3). The relatively large value indicates small rate dependence and suggests a domain structure with broad and mobile domain walls, being the preferred sites for the nucleation.

This thesis investigates both experimentally and numerically the oxidation, sintering, melting and solidification processes of different nanoparticles under various thermodynamic scenarios, with a background for energy applications. Two sets of main techniques are adopted in this work, which are isoconvensional kinetic analysis and molecular dynamics simulation. Based on the techniques of simultaneous Thermogravimetric Analysis (TGA) and Differential Scanning Calorimetry (DSC), for first time the isoconvensional kinetic analysis is applied to study the oxidation of nickel and tin nanoparticles. This method is demonstrated capable of modelling one-step nanoscale oxidation and revealing underling kinetic mechanisms. Moreover, some distinct features of nanoparticle oxidation compared with their bulk counterparts are found such as melting depression, oxidation kinetic change in the vicinity of Curie point of nickel and pressure-related two-step oxidation of tin nanoparticles. The detailed study from Molecular Dynamics (MD) simulation establishes a three-stage sintering process of two nickel nanoparticles, which is unable to be described by bulk continuum-level models. MD is applied to study the interaction between nickel and aluminium and its consequent thermo-mechanical and structural property evolution in a nickel-coating aluminium particle in a heating and cooling cycle. The simulation successfully predicts the atomic diffusion during melting and the formation of glass and crystal phases, and allows for the estimation of interior core-shell pressure. Reactive MD is then applied to simulate the oxidation of silicon nanoparticles. It predicts well the exothermal reaction process and experimentally reveals the oxygen exchange process.

The degradation of polyurethane (PU) coil coatings were studied with step scan phase modulation photo-acoustic (SS-PM-PA) FTIR, confocal Raman mapping (CRM) and scanning electron microscopy (SEM). PU coatings were oven cured for 30 seconds to reach a peak metal temperature of 232°C. The cured coatings were exposed in a QUV A accelerated ageing test with exposure time intervals of 1200 hours and 4098 hours. Isophorone diisocyanate (IPDI) cross-linker gave lower cross-linking density and degradation rate to the PU coating compared to hexamethylene diisocyanate (HDI). Cyclic trimer (CT) isocyanate cross-linker gave higher durability compared to biuret (BI). A primary amide and urea entity rich top-film was formed at the surface of degraded PU coatings, with characteristic IR bands at 1640 cm-1 and 1560 cm-1. The decomposition of allophanate in exposed HDI-CT cross-linked PU coating was indicated. The degradation of BI core produced additional urea linkage compared to allophanate. ɛ-caprolactam (Capro) blocked isocyanate gave lower cross-linking density and higher degradation rate compared to methyl ethyl ketoxime (MEKO), and 3,5 dimethyl pyrazole (DMP). The addition of melamine and HALS (less than 5%) improved the durability of PU coatings. The melamine linkage was more sensitive to the degradation compared to the urethane linkage. The higher NCO/OH resulted in more rapid degradation product build-up at the surface of the PU coating in the meantime deterred the decomposition of amide II type linkage. A FTIR peak fitting method was developed for generating degradation index plots, based on the knowledge of degradation chemistry of the PU coatings described above. The degradation rate correlation of the PU coatings exposed in the QUV A test and natural exposure sites including Liverpool, UK (LIV), Vereeniging, South Africa (SA) and Kuala Lumpur, Malaysia (KL) are demonstrated by using degradation index plot methods. The harshness of the natural exposure sites gives the order of KL > SA > LIV.

Spinal fixation systems using pedicular screws have gained popularity in manging the damaged spine. However, the loading to which individual components of a fixator are exposed are largely unknown. This thesis describes the use of a Corpectomy injury model to investigate the mechanical response of a commercial internal spinal fixator and the resultant loads acting on its rods and screws, under four separatelo ading regimens. The fixator was instrumentedw ith strain gaugesa nd tested using specially designed jigs. The results were then compared to theoretical models and any differences highlighted. An evaluation was also performed on a range of transpedicular screw designs under tensile loads. An increase in the tightening torque of the fixator clamps, ranging from 5 to 15Nm, and the inclusion of transverse elements across its vertical rods produced a combined increase in overall torsional rigidity of 89%. However, no such changes were found under axial compression and both simulated flexion and extension tests. The relative ineffectivenesso f the transversee lementsu nder sagittal loads was probably due to their spatial relationship with the fixator. The results from the instrumented fixator indicated several load response pathways, as predicted by the theoretical analysis. These pathways were influenced by several factors including, the screw angulation, the boundary conditions of the test and the addition of the transverse elements. Clamp design was critical in minimising rotational slippage of both screws and transverse elements. The results from the instrumented fixator revealed that the transpedicular screws were exposed to complex loads under each of the tests. Under tensile loads, both the increasei n screw insertion depth and a decreasein screw pitch were found to be the important parameters which affect screw performance. Analysis showed the state of stress and strain along the thread was the overriding factor in the tensile performance of these screws. This work hase mphasisedth e importance of a full biornechanicale valuation of any future designs of spinal fixators.

Compression-after-impact testing is widely used to assist in the development and selection of materials for aircraft applications. Presently, there are no standard test methods in existence. The most widely used industrial tests require large specimens which are expensive to manufacture and test. The results of an experimental study of the compression-after-impact test are reported. A miniaturised testing arrangement was used to investigate the effects of specimen width, thickness and lay-up on the measured compression strength of undamaged and impact damaged specimens. A toughened carbon I epoxy was used for the above work. In addition three other materials were tested (a carbon / polyetheretherketone (APC), a glass I epoxy (GRP) and another carbon / epoxy). The in-plane extent of delamination damage after impact was measured using an ultrasonic C-scanning method. The carbon and glass reinforced epoxy materials had similar resistance to the initiation and propagation of impact damage. The APC was much more resistant to the formation of impact damage. The measured strength of undamaged specimens was dependent upon specimen geometry, decreasing with width increase and increasing with thickness increase. The strength of impact damaged specimens was independent of width. Increasing the thickness increased the incident impact energy required to initiate damage and, therefore, delayed the onset of residual strength reductions. The strength of undamaged quasi-isotropic and 0/90 laminates was very similar and higher than for ±45 laminates. After impact the 0/90 material was strongest. The residual strength of the quasi-isotropic and ±45 materials were very similar. The APC retained the highest proportion of its initial strength over a range of incident impact energies. This was attributed to its resistance to the formation of impact damage. The GRP was the most damage tolerant material.

In materials science, the traditional methodological framework is the identification of the composition-processing-structure-property causal pathways that link hierarchical structure to properties. However, all the properties of materials can be derived ultimately from structure and bonding, and so the properties of a material are interrelated to varying degrees. The work presented in this thesis, employed artificial neural networks (ANNs) to explore the correlations of different material properties with several examples in different fields. Those including 1) to verify and quantify known correlations between physical parameters and solid solubility of alloy systems, which were first discovered by Hume-Rothery in the 1930s. 2) To explore unknown crossproperty correlations without investigating complicated structure-property relationships, which is exemplified by i) predicting structural stability of perovskites from bond-valence based tolerance factors tBV, and predicting formability of perovskites by using A-O and B-O bond distances; ii) correlating polarizability with other properties, such as first ionization potential, melting point, heat of vaporization and specific heat capacity. 3) In the process of discovering unanticipated relationships between combination of properties of materials, ANNs were also found to be useful for highlighting unusual data points in handbooks, tables and databases that deserve to have their veracity inspected. By applying this method, massive errors in handbooks were found, and a systematic, intelligent and potentially automatic method to detect errors in handbooks is thus developed. Through presenting these four distinct examples from three aspects of ANN capability, different ways that ANNs can contribute to progress in materials science has been explored. These approaches are novel and deserve to be pursued as part of the newer methodologies that are beginning to underpin material research.

This thesis comprises studies of size effects in the plasticity of metals and ceramics at length scales of the order of micrometres and includes both experimental work and theoretical development. Experimental results are presented for foil flexure (nickel and copper)and nanoindentation (ceramics and hard metals).These studies were conducted because existing data does not cover a range broad enough or with sufficient precision to test various theories. With the developed bending technique more accurate data is obtained covering a wide range of strain, especially around the key region of the elastic-plastic transition. Moreover, the interaction between grain and thickness size effect is successfully studied by varying the ratio of grain size over thickness of the foils. After carefully calibrating the indenters, the macroscopic indentation yield strength for ceramics and high strength metals is determined in a direct way by using spherical nanoindentation. The magnitude of size effect is significantly different between metals and ceramics. By comparing the Berkovich and spherical indentation size effect, the results implies that the contact size, a, is the most fundamental length scale in the indentation size effect, independent of the indenter shape. The indentation strength is found to be inversely scaled with the square root of a. The slip-distance theory (based on (Conrad et al, 1967)) with an effective length scale reconciling intrinsic and extrinsic size effects appears able to account for the size effects in all contexts, without requiring strain gradient plasticity theory or an implicit characteristic length.

This thesis aims to explore the effect of pre-strain on the viscoelastic behaviour of rubber materials. There are various conflicting theories in the literature regarding the strain dependence and resulting anisotropy of the viscoelastic behaviour. This thesis seeks to measure the behaviour and to study the possibility of using Finite Element Analysis (FEA) to predict the static behaviour of a rubber cylinder in combined torsion-tension and also the viscoelastic behaviour of rubber under various complex loadings using a Bergstrom-Boyce model1-4. To measure the induced anisotropy, a rubber test piece is subjected to a simple extension l and then it is subjected to small strain oscillations in the direction of the pre-extension or in shear. These two different deformations will allow the extent of the anisotropy in the viscoelastic behaviour induced by the pre-extension to be measured. Kuhn and Kunzle5 found that the loss factor resulting from a small oscillation decreased as a function of the static pre-strain. They and many others have interpreted this as a lowering of internal viscosity due to chain orientation. However, a simple analysis shows that this effect is due to geometric changes alone and that the essential viscoelastic behaviour expressed in terms of the deformed dimension after the application of the pre-strain as the loss modulus for an unfilled rubber is constant with strain up to an extension ratio of 2. It is also isotropic in behaviour for filled rubber compounds such as carbon black. For fumed silica filled rubber, the picture is more complex. For a moderately carbon black (25 phr) filled rubber, the loss modulus is still independent of the pre-strain for normal working strains but at highly filler contents (above 50 phr), the loss modulus increases with pre-strain at extension ratios somewhat less than 2. With silica, the coupling agent dominates the viscoelastic behaviour. For filled rubber, the change in loss modulus with strain can in part be explained by strain amplification, slippage of rubber around the filler, and shape factor effects. This approach can help to further understand the mechanism of filler reinforcement in rubber materials. Another complex loading is also used to validate these results with a static pure shear superimposed with simple shear oscillation. The results confirm the loss modulus is independent of the pre-strain for unfilled rubber and lightly filled rubber but for the most highly filled rubber, the test is unsuitable as the smallest oscillating strains were too great for linear viscoelastic behaviour. The Finite Element Analysis (FEA) shows that a rubber cylinder in combined torsion-tension test can be modelled accurately as an elastic component provided that the appropriate strain energy function (SEF) and geometry are used in the model. The correct torque and the second order effect whereby a reduction in the axial force resulting from the torsion of a pre-strained rod can both be accurately represented. The viscoelastic behaviour under various complex loadings was modelled using the Bergstrom and Boyce model1. The results show that this model can predict behaviour for uniaxial but in a complex loading the model was inappropriate.

Composite integral armour plays an important role in future combat system. Despite numerous experimental studies there are still disadvantages such as complex manufacturing process, relatively big damage area, difficult to repair and limit shape etc. Composite integral armour without all these problems is essential for the success of future main battle tank which has a total weight of only 20 tons. 3D fabrics are seen as potential solution to poor impact damage tolerance of textile composites. Binder yarns in through-thickness direction can bridge cracks and stop crack tip growth resulting very good impact damage tolerance. The major purple of this work is to incorporate new materials and new configuration into composite integral armour. The underlying premise is that ballistic performance of new armour is judged mainly by single hit ballistic limit followed by damage resistance which in turn followed by energy absorption in high energy low velocity impact. Computer simulation of 3D textile composites and damage mechanism study were used through-out the study for analysing and explaining experimental results. Judged by these properties, conclusions regarding to ballistic performance of eight 3D texile composties were made. The benefit of the work will be a new explanation of composite armour research. This will help the success of future combat system.

The conventional materials discovery is a kind of empirical (“trial and error”) science that of handling one sample at a time in the processes of synthesis and characterization. However, combinatorial methodologies present the possibility of a vastly increased rate of discovery of novel materials which will require a great deal of conventional laboratory work. The work presented in this thesis, involved the practice of a conceptual framework of combinatorial research on BaTiO3-based positive temperature coefficient resistor (PTCR) materials. Those including (i) fabrication of green BaTiO3 base discs via high-throughput dip-pen printing method. Preparation and formulation of BaTiO3 inks (selection of dispersant and binder/volume fraction) were studied. The shape of drying residues and the morphogenesis control of droplet drying were discussed. (ii) investigation of a fast droplet-doping method, which induced the dopant precursor solution infiltrating into the porous BT base disc. Various characterization methods were used to examine the dopant distribution in the body of disc. (iii) devising a high-throughput electrical measurement system including an integrated unit of temperature control and automatic measurement operation, and an arrayed multichannel jig. (iv) synthesis of donor-doped BaTiO3 libraries, which involved lanthanum, erbium, yttrium as donor elements and manganese as an acceptor dopant element respectively. Their temperature dependant resistivities were also explored. The work successfully developed an integrated tool including high-throughput synthesis of a large batch of libraries and high-throughput electrical property measurement for combinatorial research on BaTiO3-based PTCR ceramics. The Abstract ii combinatorial method, thus validated, has the potential to deliver dopant-doped BTbased PTCR libraries rapidly with a very wide range of dopant mixtures and concentrations for electrical property measurement and deserves to be applied to other low level dopant ceramic systems. These approaches are novel and paving the way for other new materials selection and materials research.